CVPR 2026 论文列表

Contrastive vision-language (V&L) models remain a popular choice for various applications. However, several limitations have emerged, most notably the limited ability of V&L models to learn compositional representations. Prior methods often addressed this limitation by generating custom training data to obtain hard-negative samples. Hard-negatives have been shown to improve performance on compositionality tasks, but are often specific to a single benchmark, do not generalize, and can cause substantial degradation of basic V&L capabilities such as zero-shot or retrieval performance, rendering them impractical. In this work we follow a different approach. We identify two root causes that limit compositionality performance of V&Ls: 1) Long training captions do not require a compositional representation; and 2) The final global pooling in the text and image encoders leads to a complete loss of the necessary information to learn binding in the first place. As a remedy, we propose two simple solutions: 1) We obtain short concept centric caption parts using standard NLP software and align those with the image; and 2) We introduce a parameter-free cross-modal attention-pooling to obtain concept centric visual embeddings from the image encoder. With these two changes and simple auxiliary contrastive losses, we obtain SOTA performance on standard compositionality benchmarks, while maintaining or improving strong zero-shot and retrieval capabilities. This is achieved without increasing inference cost. We release the code for this work at https://github.com/SamsungLabs/concept_centric_clip.

Crime anticipation enables proactive public safety interventions, yet existing video security systems remain largely reactive, unable to detect precursors of crime. While current visual language models (VLM)-based video understanding methods show promise in high-level reasoning, they are not designed to explicitly model the spatio-temporal causal relationships essential for anticipating crimes. We address this limitation by two causal-driven components. First, we develop the Spatio-Temporal Causal Reasoning Crime (STCRC) dataset, a hierarchical dataset comprising 73K samples across five progressive causal reasoning tasks, facilitating criminal precursors learning. Second, we propose the Spatio-Temporal Causal Hypergraph (STCH), a streaming module that transforms implicit entity dynamics into explicit causal structures to enhance causal reasoning for crime in VLMs. By combining these two components, our framework advances real-time crime anticipation, achieving improvements in anticipatory tasks: a 70.7% relative improvement in crime classification, a 10.1% in crime detection, and a 3.7% reduction in temporal prediction error.

Open-vocabulary part segmentation (OVPS) aims to segment objects into fine-grained parts while generalizing to unseen categories. Existing VLM-based methods face two challenges: (1) object over-segmentation, caused by overly broad semantic activations, and (2) part under-segmentation, resulting from weak fine-grained perception. To address these issues, we propose HOPS, a two-stage framework for hierarchical open-vocabulary part segmentation. HOPS introduces a bidirectional semantic-structural attention fusion mechanism that integrates CLIP's semantic alignment with DINO's structural perception. In the object segmentation stage, the Attention-Aware Filtering Module (AFM) refines cross-modal similarity maps via semantic-structural attention to suppress object over-segmentation. In the part segmentation stage, the Affinity-Guided Enhancement Module (AEM) iteratively propagates part responses to progressively expand activation regions, effectively mitigating part under-segmentation. Experiments on Pascal-Part-116, ADE20K-Part-234, and PartImageNet demonstrate that HOPS achieves state-of-the-art performance with superior generalization. Our code is available at https://github.com/TJU-IDVLab/HOPS.

Large vision-language models (LVLMs) are powerful, yet they remain unreliable due to object hallucinations. In this work, we show that in many hallucinatory predictions the LVLM effectively ignores the image and instead relies on previously generated output ("prelim") tokens to infer new objects. We quantify this behavior via the mutual information between the image and the predicted object conditioned on the prelim, demonstrating that weak image dependence strongly correlates with hallucination. Building on this finding, we introduce the Prelim Attention Score (PAS), a lightweight, training-free signal computed from attention weights over prelim tokens. PAS requires no additional forward passes and can be computed on the fly during inference. Exploiting this previously overlooked signal, PAS achieves state-of-the-art object-hallucination detection across multiple models and datasets, enabling real-time filtering and intervention.

Vision-Language-Action (VLA) models have recently achieved remarkable progress in robotic manipulation, yet they remain limited in failure diagnosis and learning from failures. Additionally, existing failure datasets are mostly generated programmatically in simulation, which limits their generalization to the real world. In light of these, we introduce ViFailback, a framework designed to diagnose robotic manipulation failures and provide both textual and visual correction guidance. Our framework utilizes explicit visual symbols to enhance annotation efficiency. We further release the ViFailback dataset, a large-scale collection of 58,128 Visual Question Answering (VQA) pairs along with their corresponding 5,202 real-world manipulation trajectories. Based on the dataset, we establish ViFailback-Bench, a benchmark of 11 fine-grained VQA tasks designed to assess the failure diagnosis and correction abilities of Vision-Language Models (VLMs), featuring ViFailback-Bench Lite for closed-ended and ViFailback-Bench Hard for open-ended evaluation. To demonstrate the effectiveness of our framework, we built the ViFailback-8B VLM, which not only achieves significant overall performance improvement on ViFailback-Bench but also generates visual symbols for corrective action guidance. Finally, by integrating ViFailback-8B with a VLA model, we conduct real-world robotic experiments demonstrating its ability to assist the VLA model in recovering from failures. Project page: https://x1nyuzhou.github.io/vifailback.github.io/

Histopathology whole slide images (WSIs) are gigapixel images that present significant challenges in generating effective representations that capture both local histological features and their global spatial organization. Current pathology foundation models focus primarily on local patch-level features while neglecting the complex spatial relationships that pathologists rely on for diagnosis and prognosis. We introduce TopoSlide, a novel self-supervised representation learning framework that leverages persistent homology from topological data analysis to capture the global spatial organization of tissue architecture in WSIs. Our method decomposes slides into histologically meaningful clusters using patch-level embeddings, then characterizes their spatial arrangement through topological descriptors. We train a vision transformer to predict cluster topology from slide-level embeddings using a conditional multi-task objective that integrates local patch features with their topological attributes. Evaluated across lung adenocarcinoma and breast cancer cohorts, TopoSlide achieves superior performance improving histologic pattern retrieval by up to 15% in majority voting macro F1 score, and competitive survival and gene mutation predictions, while training on only hundreds of slides compared to hundreds of thousands for foundation models. Our results demonstrate that topology-aware learning provides a powerful inductive bias for pathology representation learning, enabling both improved performance and novel topology-based conditional retrieval capabilities for clinical applications. Our code and models are publicly available.

Recent progress in video-to-video (V2V) translation has enabled realistic resimulation of embodied AI demonstrations, a capability that allows pretrained robot policies to be transferable to new environments without additional data collection. However, prior works can only operate on a single view at a time, while embodied AI tasks are commonly captured from multiple synchronized cameras to support policy learning. Naively applying single-view models independently to each camera leads to inconsistent appearance across views, and standard transformer architectures do not scale to multi-view settings due to the quadratic cost of cross-view attention. We present VideoWeaver, the first multimodal multi-view V2V translation framework. VideoWeaver is initially trained as a single-view flow-based V2V model. To achieve an extension to the multi-view regime, we propose to ground all views in a shared 4D latent space derived from a feed-forward spatial foundation model, namely, Pi3. This encourages view-consistent appearance even under wide baselines and dynamic camera motion. To scale beyond a fixed number of cameras, we train views at distinct diffusion timesteps, enabling the model to learn both joint and conditional view distributions. This in turn allows autoregressive synthesis of new viewpoints conditioned on existing ones. Experiments show superior or similar performance to the state-of-the-art on the single-view translation benchmarks and, for the first time, physically and stylistically consistent multi-view translations, including challenging egocentric and heterogeneous-camera setups central to world randomization for robot learning.

The latest progress in text-to-3D generative models makes it possible to generate high-quality 3D content. Recent text-to-3D large model have achieved remarkable breakthroughs in multi-view consistency. However, their effectiveness is often affected by inherent biases, resulting in sensitivity to design settings such as prompt format, leading to difficulty understanding complex prompts. To help text-to3D generative models understand more diverse prompts, we propose a framework to localize and mitigate the bias in the current text-to-3D large model. Specifically, we first use the existing model to generate 3D content and use the quality evaluation model to identify the cross-modality bias. Then, we use the predicted quality score to quantify the contribution of the prompt text to the bias. Finally, in order to reduce these biases, we construct diverse pairwise examples to help the current text-to-3D large model construct unbiased visual-text connections. The experiment shows that our method has achieved competitive results and can provide higher quality, more diverse 3D content compared to existing methods.

Learning-based video quality assessment (VQA) has advanced rapidly, yet progress is increasingly constrained by a disconnect between model design and dataset curation. Model-centric approaches often iterate on fixed benchmarks, while data-centric efforts collect new human labels without systematically targeting the weaknesses of existing VQA models. Here, we describe MDS-VQA, a model-informed data selection mechanism for curating unlabeled videos that are both difficult for the base VQA model and diverse in content. Difficulty is estimated by a failure predictor trained with a ranking objective, and diversity is measured using deep semantic video features, with a greedy procedure balancing the two under a constrained labeling budget. Experiments across multiple VQA datasets and models demonstrate that MDS-VQA identifies diverse, challenging samples that are particularly informative for active fine-tuning. With only a 5% selected subset per target domain, the fine-tuned model improves mean SRCC from 0.651 to 0.722 and achieves the top gMAD rank, indicating strong adaptation and generalization.

Recently, Spectral Compressive Imaging (SCI) has achieved remarkable success, unlocking significant potential for dynamic spectral vision. However, existing reconstruction methods, primarily image-based, suffer from two limitations: (i) Encoding process masks spatial-spectral features, leading to uncertainty in reconstructing missing information from single compressed measurements, and (ii) The frame-by-frame reconstruction paradigm fails to ensure temporal consistency, which is crucial in the video perception. To address these challenges, this paper seeks to advance spectral reconstruction from the image level to the video level, leveraging the complementary features and temporal continuity across adjacent frames in dynamic scenes. Initially, we construct the first high-quality dynamic hyperspectral image dataset (DynaSpec), comprising 30 sequences obtained through frame-scanning acquisition. Subsequently, we propose the Propagation-Guided Spectral Video Reconstruction Transformer (PG-SVRT), which employs a spatial-then-temporal attention to effectively reconstruct spectral features from abundant video information, while using a bridged token to reduce computational complexity. Finally, we conduct simulation experiments to assess the performance of four SCI systems, and construct a DD-CASSI prototype for real-world data collection and benchmarking. Extensive experiments demonstrate that PG-SVRT achieves superior performance in reconstruction quality, spectral fidelity, and temporal consistency, while maintaining minimal FLOPs.

Dyadic interactive head generation aims to synthesize realistic head motions that respond both verbally and non-verbally to an interlocutor in real-time conversation. The existing works often focus on offline scenarios, and struggle with a shallow understanding of the multimodal conversational context while also lacking long-term coherence. To address these limitations, we propose MimicTalker, a novel method for producing real-time, contextually-aware, and long-term consistent interactive head motions. To this end, we propose a Multimodal Interactive Context Extraction (MICE) module to capture both instantaneous and long-term multimodal interactive information from the interlocutor. To enhance in-depth conversational understanding, we propose a Semantic-enhanced Dynamic Interaction (SDI) module to integrate the intentions and topics of the conversation, which are automatically extracted through an LLM-based analyzer. Further, we propose a semantic-guided Motion Style Memory (MSM) mechanism, enabling the long-term motion consistency throughout the conversation. We conduct experiments on both short conversational segments (25 seconds) and extended dialogues (6 minutes), and the comprehensive experiments demonstrate that our method significantly outperforms existing approaches.

Generative data augmentation has created new opportunities for improving image recognition models. Despite these advances, most existing augmentation methods process images holistically, without accounting for the uneven distribution of discriminative information in classification tasks, where foreground subjects typically contain richer category-relevant signals than background context. Ignoring this imbalance can introduce unintended semantic shifts in generated samples, thereby weakening the model's ability to capture the intrinsic ontology of the target object. In fact, human perception operates in a similar manner: subject identity remains stable, while contextual variations are naturally tolerated as long as overall coherence is preserved. Inspired by this observation, we propose OntoAug, an ontology-oriented data augmentation framework that explicitly distinguishes between subject and environment. OntoAug separates foreground and background through structured layout control and guides diffusion models to generate samples with consistent subjects and diverse contextual variations. By explicitly introducing foreground masks and diverse background prompts during the generation process, the framework enhances both semantic fidelity and diversity in the synthesized data. Extensive experiments show that OntoAug achieves strong performance across multiple tasks, including image classification, few-shot learning, weakly supervised object localization (WSOL), and large vision-language model reasoning.

Large Vision-Language Models (VLMs) enable strong multimodal reasoning but incur heavy inference costs from redundant visual tokens. Token pruning alleviates this issue, yet existing approaches face limitations. Attention-based methods rely on raw attention scores, which are often unstable across layers and heads and can lead to redundant selections. Diversity-based methods improve robustness by selecting tokens far apart in feature space, but risk dropping regions needed for accurate prediction. We propose ZOO-Prune, a training-free framework built on the intuition that highly sensitive tokens have a stronger influence on the model's output and capture complementary visual cues rather than redundant ones. To achieve this, we estimate token sensitivity using zeroth-order perturbations at the lightweight projection layer. This measures how small random perturbations affect the projected features and enables efficient approximation of each token's influence without backpropagation. Extensive experiments across multiple VLMs and benchmarks show that ZOO-Prune consistently outperforms prior methods while pruning up to 94.4% of tokens without sacrificing accuracy. Our method also improves efficiency, reaching up to 2.30x faster end-to-end inference compared to the baseline. Code is available at https://aim-skku.github.io/ZOO-Prune.

Recent advances in diffusion models and vision-language models (VLMs) have significantly enhanced the controllability of image editing. Methods like FlowEdit enable step-by-step editing along a visible, noise-free trajectory, where each intermediate result is a clear image, eliminating the need for full noise inversion. However, these approaches still operate in pixel space or VAE latent space, where intermediate outputs often suffer from visual artifacts, distortions, or unrealistic details--making reliable semantic evaluation difficult. Furthermore, they remain open-loop systems, applying static edits without feedback to guide or correct the editing process adaptively. We propose UniEdit-I, the first training-free, closed-loop image editing framework that operates entirely within the semantic latent space of a unified VLM by introducing an Understanding-Editing-Verifying (UEV) loop:(1) Understanding: parses the source image and editing instruction into a structured source prompt and a minimal target specification;(2) Editing: applies dynamic semantic offsets, with a configurable feedback weighting mechanism that adaptively modulates editing intensity based on real-time alignment feedback;(3) Verifying: leverages the VLM's own multimodal reasoning capability to evaluate the intermediate output along multiple semantic dimensions and trigger early stopping or refinement. By transforming the VLM from a post-hoc evaluator into an in-process conductor, UniEdit-I establishes the first semantics-driven, self-correcting closed-loop image editing pipeline. Evaluated on GEdit-Bench, UniEdit-I achieves state-of-the-art performance without any fine-tuning or architectural modifications, and even surpasses several large-scale pre-trained editors.

This paper presents a multi-agent perception-action exploration alliance, dubbed A4VL, for efficient long-video reasoning. A4VL operates in a multi-round perception-action exploration loop with a selection of VLM agents. In each round, the team of agents performs video question-answer (VideoQA) via perception exploration followed by action exploration. During perception exploration, each agent learns to extract query-specific perception clue(s) from a few sampled frames and performs clue-based alignment to find the video block(s) that are most relevant to the query-specific event. During action exploration, A4VL performs video reasoning in three steps: (1) each agent produces its initial answer with rational, (2) all agents collaboratively scores one another through cross-reviews and relevance ranking, and (3) based on whether a satisfactory consensus is reached, the decision is made either to start a new round of perception-action deliberation by pruning (e.g., filtering out the lowest performing agent) and re-staging (e.g., new-clue and matching block based perception-action exploration), or to conclude by producing its final answer. The integration of the multi-agent alliance through multi-round perception-action exploration, coupled with event-driven partitioning and cue-guided block alignment, enables A4VL to effectively scale to real world long videos while preserving high quality video reasoning. Evaluation Results on five popular VideoQA benchmarks show that A4VL outperforms 18 existing representative VLMs and 11 recent methods optimized for long-video reasoning, while achieving significantly lower inference latency. Our code is released at https://github.com/git-disl/A4VL.

Image-to-point-cloud (I2P) registration aims to align 2D images with 3D point clouds by establishing reliable 2D-3D correspondences. The drastic modality gap between images and point clouds makes it challenging to learn features that are both discriminative and generalizable, leading to severe performance drops in unseen scenarios. We address this challenge by introducing a heterogeneous graph that enables refining both cross-modal features and correspondences within a unified architecture. The proposed graph represents a mapping between segmented 2D and 3D regions, which enhances cross-modal feature interaction and thus improves feature discriminability. In addition, modeling the consistency among vertices and edges within the graph enables pruning of unreliable correspondences. Building on these insights, we propose a heterogeneous graph embedded I2P registration method, termed Hg-I2P. It learns a heterogeneous graph by mining multi-path feature relationships, adapts features under the guidance of heterogeneous edges, and prunes correspondences using graph-based projection consistency. Experiments on six indoor and outdoor benchmarks under cross-domain setups demonstrate that Hg-I2P significantly outperforms existing methods in both generalization and accuracy. Code is released on https://github.com/anpei96/hg-i2p-demo.

Vision-Language Models (VLMs) have emerged as versatile solutions for zero-shot question answering (QA) across various domains. However, enabling VLMs to effectively comprehend structured graphs and perform accurate, efficient QA remains challenging. Existing approaches typically rely on a single type of graph topology representation (GTR) of graphs, such as fixed-style visual images or unified text descriptions. This "one-size-fits-all" strategy often neglects model-specific and task-specific preferences, resulting in inaccurate or overly lengthy responses to graph-related queries. To address this, we propose the DynamicGTR framework, which dynamically selects the optimal GTR for each query during inference, thereby enhancing the zero-shot graph QA capabilities of VLMs with a customizable accuracy and brevity trade-off. Extensive experiments show that DynamicGTR not only improves VLM-based graph algorithm QA performance but also successfully transfers the experience trained from synthetic graph algorithm tasks to real-world applications like link prediction and node classification, without any additional training. Additionally, DynamicGTR demonstrates strong transferability across tasks, domains, and models, suggesting its potential as a flexible solution for broad graph scenarios.

Multimodal contrastive learning typically relies on pairwise similarities for alignment, but recent work has shown that Gramian volumes can capture higher-order correlations across modalities. However, Euclidean Gramian volumes suffer from volume collapse under L2 normalization, concentrating near unity with minimal discriminative variance. Hyperbolic geometry's exponential volume growth naturally addresses this via variance preservation, motivating us to extend Gramian alignment to hyperbolic space. Yet preliminary experiments reveal that pure hyperbolic geometry alone is insufficient: while it preserves variance, it underperforms Euclidean baselines on cross-category discrimination. We introduce HyperGRAM, a hybrid geometry framework that combines Euclidean discriminative stability with hyperbolic semantic variance through learnable mixing. Using the numerically stable Lorentz model, HyperGRAM enables volumes to serve dual roles: discriminating matched from mismatched triplets while preserving semantic sensitivity within matched pairs that reflects interpretation spaces (the set of valid multimodal realizations). Evaluation across four video-text benchmarks demonstrates that hybrid geometry consistently outperforms both pure Euclidean and pure hyperbolic variants, achieving significant zero-shot improvements with cross-dataset semantic sensitivity exhibiting contrasting correlation patterns.

Vision Transformers (ViTs) achieve state-of-the-art performance on a wide range of vision tasks, yet they remain highly vulnerable to adversarial perturbations due to the lack of explicit region-level semantic modeling. Adversarial perturbations are typically local and spatially structured, whereas the globally coupled self-attention and spatially uniform feed-forward networks in ViTs propagate local corruptions across the whole image without enforcing consistency within semantically coherent regions. To mitigate this mismatch, we propose Region-aware Mixture-of-Experts, namely "ReMoE", a plug-and-play module that replaces the standard feed-forward network (FFN) with a region-aware expert layer. Specifically, our ReMoE strategically introduces multi-granularity experts (i.e., global, center, and regional) and couples them with an attention-guided routing mechanism that operates on patch-to-region (P2R) and region-to-patch (R2P) transformations. This mechanism adaptively activates the most relevant experts for each spatial location according to its attention profile, enabling the model to capture region-level semantics and local context while preserving global consistency, thereby providing a stronger inductive bias for adversarially robust ViT representations. Extensive experiments demonstrate that our ReMoE substantially improves the adversarial robustness of ViTs with only marginal additional computational cost. Our code is available at https://github.com/zhongskr0114/ReMoE.